Why did you choose to join the faculty in the Department of Biostatistics and Bioinformatics?
I was drawn to Duke because of its deeply collaborative culture and its bold institutional commitment to advancing artificial intelligence in biomedical discovery. Duke’s recently launched Discovery AI initiative reflects a clear vision to integrate AI and machine learning into fundamental biological research while providing shared computational infrastructure and cutting-edge technologies. The Department of Biostatistics and Bioinformatics at Duke School of Medicine is uniquely positioned at the interface of statistical innovation, data science, and translational medicine. I saw an exceptional opportunity to contribute to and help shape an environment where methodological advances in AI can directly accelerate our understanding of human biology and disease. Joining Duke allows me to collaborate across disciplines and build next-generation computational frameworks that transform large-scale biomedical data into actionable knowledge.
Where were you working previous to Duke? What was your role there?
Prior to joining Duke, I was a tenure-track Assistant Professor in the Department of Biostatistics at Columbia University and an affiliated member of the Data Science Institute. There, I led a research group focused on developing statistical machine learning and foundation-model approaches for high-dimensional genomic data, including single-cell and spatial multi-omics.
My work centered on bridging methodological rigor with biological insight, such as designing scalable AI systems that enable researchers to decode gene regulation, characterize cellular heterogeneity, and predict regulatory responses across tissues and disease contexts.
What is your approach to mentorship/teaching?
My mentorship philosophy is grounded in cultivating independence, intellectual curiosity, and scientific courage. I strive to create an environment where trainees feel empowered to pursue ambitious questions while developing strong quantitative foundations. Because modern biomedical science increasingly relies on AI-driven discovery, I emphasize interdisciplinary training by helping students become fluent in statistics, machine learning, and biological reasoning. Equally important, I encourage collaborative thinking and ethical responsibility so that emerging AI technologies are applied thoughtfully and for meaningful societal benefit. Ultimately, my goal is to mentor scholars who will not only advance methodology but also redefine how computation shapes the future of medicine.
What will you be teaching/what types of learners will you be working with?
I look forward to teaching graduate-level courses in statistical machine learning, AI for biomedical data science, and novel applications of large language models. My learners will include PhD students, master’s students, postdoctoral fellows, and interdisciplinary trainees from fields such as AI, genomics, computational biology, and biomedical engineering. In the classroom, I aim to connect theory with practice, equipping students to develop principled algorithms while working effectively with complex real-world datasets. Given the rapid evolution of AI, I am particularly excited to prepare trainees to lead the next generation of data-driven biomedical research.
What are your research interests?
My research focuses on developing next-generation artificial intelligence and statistical methods to understand gene regulation and cellular dynamics at scale. I am particularly interested in creating large, transferable systems that can learn from multimodal data such as gene expression, chromatin accessibility, and spatial genomic measurements.
By integrating AI with rigorous statistical modeling, my group seeks to:
* Predict regulatory responses across cell types and conditions
* Reveal mechanisms underlying human disease
* Enable more precise and data-driven therapeutic strategies.
More broadly, I am motivated by the vision of transforming AI into a discovery engine for biomedicine, such as generating hypotheses, guiding experiments, and accelerating translation from data to therapeutic understanding.
Are there any major research grants you have been a part of or are currently working on?
My research program has been supported by federal funding dedicated to advancing computational genomics and AI-enabled biomedical discovery, including an NIH Maximizing Investigators’ Research Award (MIRA) and other NIH-supported projects focused on gene regulatory modeling and multimodal data integration. I am currently leading efforts to develop foundation models that predict regulatory responses in single cells and integrate diverse genomic modalities to better understand disease biology. These projects bring together collaborators across statistics, computer science, and medicine and reflect my commitment to building scalable computational frameworks that empower the broader scientific community.
Is there any other information you would like to share?
I am thrilled to join Duke at a pivotal moment as the university expands its leadership in artificial intelligence through major interdisciplinary initiatives. Duke’s commitment to collaborative, impact-driven science makes it an inspiring place to pursue ambitious research at the intersection of AI and human health. Beyond advancing methodology, I am passionate about mentoring the next generation of scientists, building inclusive research communities, and developing computational tools that meaningfully improve our understanding of biology and disease. I look forward to forming new collaborations across the School of Medicine and the university and contributing to a community that is shaping the future of AI in biomedicine.